Layer dynamic aggregation and linkage visualization method and system based on context awareness

Abstract

The invention provides a layer dynamic aggregation and linkage visualization method and system based on context awareness, and the method comprises the steps: 1, constructing a master-slave layer linkage model, which comprises the following steps: appointing a master layer and at least one slave layer; constructing a linkage rule between the master layer and the slave layer; defining a data aggregation operation, wherein the data aggregation operation is used for performing aggregation calculation on the selected slave layer point elements; defining a visualization mode of a data aggregation result; 2, capturing a user interaction event and a context; step 3, dynamic aggregation calculation; and step 4, visualizing an aggregation result according to a visualization mode. According to the method and the device, instant and automatic layer data dynamic aggregation and multi-view linkage visualization based on the user interaction context are realized, so that the user experience and the decision analysis efficiency are improved.

Description

Technical Field

[0001] This invention belongs to the fields of geographic information systems, real-time big data computing and interactive data visualization technology, and in particular relates to a method and system for dynamic aggregation and linkage visualization of layers based on context awareness. Background Technology

[0002] Geographic Information Systems (GIS) play a crucial role in various current information systems for government decision-making, development planning, and business intelligence. GIS systems can overlay data from different sources (such as population, facilities, and economic indicators) onto a map as independent "layers." To analyze the spatial relationships between layers, for example, to analyze the "population count" (point features in another layer) within the service area of ​​a "park" (a polygonal feature in one layer), existing technologies typically employ the following methods: 1. Explicit Spatial Analysis Toolflow: Users need to actively invoke a series of GIS analysis tools. For example, first, the "Buffer Analysis" tool is used to generate a buffer polygon with a specified radius (e.g., 500 meters) centered on the selected park; then, the "Spatial Overlay Analysis" or "Spatial Connection" tools are used to perform calculations between the buffer and the population point layer, filtering out population points falling within the buffer; finally, the user needs to open the attribute table of the filtered results and manually count or summarize them to obtain the final population count. 2. Static Layer Query and Filtering: Some systems allow users to click on a feature on the map and then use preset query functions to filter and highlight features in another layer that meet spatial criteria (such as "contained within..."). However, this method usually only involves "filtering" and "highlighting" features and does not perform real-time data "aggregation" (such as summation, counting, and averaging). Users still need additional steps to obtain quantified statistical results.

[0003] While these existing technologies can achieve the functions, their inherent defects become increasingly prominent when faced with massive data and high-frequency interaction scenarios: (1) Cumbersome operation and low efficiency: Users need to have professional GIS operation knowledge and go through multiple steps to complete a simple correlation analysis. The whole process takes a long time and cannot meet the needs of real-time decision-making. (2) Lack of immediacy and context awareness: The system passively responds to the user's tool calls and cannot intelligently "perceive" the user's analysis intention (i.e., context) based on a simple interactive action (such as a click) on the main layer, and actively and in real time complete the aggregation calculation and result presentation of the subordinate layer data. (3) Fragmented result presentation and lack of linkage: The analysis results (such as a temporary buffer layer or an independent attribute table) are separated from the user's original interactive view. Data and status cannot be updated in real time and synchronously between multiple views such as maps, charts, and information panels, and the visualization effect is not intuitive. (4) High resource consumption: Explicit analysis tools often generate temporary, static intermediate data files or layers. Doing this for every query will cause unnecessary waste of disk and memory resources, especially in high-concurrency scenarios, where system performance will drop sharply. Summary of the Invention

[0004] The purpose of this invention is to provide a context-aware method and system for dynamic aggregation and linked visualization of layers. This method achieves real-time, automated dynamic aggregation of layer data and multi-view linked visualization based on user interaction context, thereby greatly improving user experience and decision analysis efficiency. The technical solution adopted is as follows: A context-aware method for dynamic aggregation and linkage visualization of layers includes the following steps: Step 1: Construct a master-slave layer linkage model, including the following steps: Specify one main layer and at least one subordinate layer; Establish linkage rules between the main layer and the subordinate layers; Define a data aggregation operation, which is used to perform aggregation calculations on selected features from the layer points; Define how to visualize the results of data aggregation; Step 2: Capture user interaction events and context; The user interaction event refers to the mouse action when a user clicks on an element on the main layer; the context refers to the information of the clicked element on the main layer. Step 3, Dynamic Aggregation Calculation: Step 3A: Match linkage rules based on context; Step 3B: Filter the point features in the layer according to the linkage rules to obtain the set of features in the layer, i.e., the filtering result; Step 3C: Perform data aggregation on the filtered results to obtain the aggregated results; Step 4: Visualize the aggregation results according to the visualization method.

[0005] Preferably, the linkage rule includes one of the following: setting a distance around the clicked element in the main layer, or the topological relationship between the clicked element in the main layer and the element itself.

[0006] Preferably, the information of the clicked feature on the main layer includes: ID, geometric information, and attribute information.

[0007] Preferably, the information of the sub-layer point features in the sub-layer feature set includes: ID, attribute information, and data value.

[0008] Preferably, step 1 further includes a step of preprocessing the point features from the layer: Step 1A: Construct a memory-based spatial index; Step 1B: Construct Geohash encoding: Iterate through each point feature in the layer and obtain its latitude and longitude coordinates; Then, the Geohash algorithm is used to encode the two-dimensional latitude and longitude coordinates of each point feature into a one-dimensional string with a specified precision; Step 1C: Construct an indexed in-memory data structure: The Geohash codes of all point features and their corresponding IDs are stored in an in-memory data structure that supports prefix matching.

[0009] Preferably, step 3B specifically includes the following steps: Step 3B1: Determine the query range: Based on the event context and linkage rules, the spatial range of this query is calculated in memory; Step 3B2: Obtain the Geohash encoding prefix of the query range: Calculate the minimum bounding box of the query range; Then, based on the required query precision, the minimum bounding rectangle is decomposed into a set of Geohash-encoded prefixes that can completely cover it; Step 3B3: Filter from the collection of layer features: Using the Geohash encoding prefixes in the Geohash encoding prefix set, a query is performed in the memory index built in step 1C to retrieve all Geohash codes containing the Geohash encoding prefixes. The point features corresponding to all Geohash code IDs are thus formed from the layer feature set.

[0010] Preferably, the indexed memory data structure is a hash table (HashMap) or a prefix tree (Trie).

[0011] A context-aware layer dynamic aggregation and linkage visualization system, used to implement a context-aware layer dynamic aggregation and linkage visualization method, includes: Master-slave layer linkage model configuration module; The interactive listening module, the real-time aggregation computing engine, and the linked visualization rendering module are connected in sequence; One input of the interactive listening module is connected to the output of the main layer; One input of the real-time aggregation computing engine is connected to the output of the master-slave layer linkage model configuration module.

[0012] Compared with the prior art, the advantages of the present invention are: By introducing a "master-slave layer" linkage model, a memory-based spatial indexing and query algorithm based on Geohash encoding, and an event-driven real-time aggregation mechanism, users can instantly obtain dynamic aggregation statistics of related layers through a single simple interaction, and simultaneously visualize them in multiple views such as maps, charts, and information panels. This provides users with a "what you see is what you get" efficient and intuitive immersive data exploration experience. Attached Figure Description

[0013] Figure 1 This is a diagram of the architecture of a context-aware layer dynamic aggregation and linkage visualization system. Figure 2 This is a flowchart of a context-aware layer dynamic aggregation and linkage visualization method. Detailed Implementation

[0014] The following will describe in more detail the context-aware layer dynamic aggregation and linkage visualization method and system of the present invention with reference to the schematic diagrams, which illustrate preferred embodiments of the invention. It should be understood that those skilled in the art can modify the invention described herein while still achieving the advantageous effects of the invention. Therefore, the following description should be understood as being of general knowledge to those skilled in the art and is not intended to limit the invention.

[0015] The core idea of ​​this invention is to transform the traditional user-driven "multi-step analysis process" into a "single-step trigger-automatic processing" mode driven by the system based on preset rules and user context.

[0016] The key lies in constructing a linked "Master-Slave Layer" model. The system predefines the logical relationships between layers; for example, the "Metro Layer" is defined as the master layer, and the "Population Layer" is defined as the slave layer.

[0017] When a user interacts with the main layer (such as clicking on a park), the interaction triggers an event. The system captures this event and uses the clicked park element as "context." Using an efficient in-memory spatial indexing algorithm, it performs spatial filtering and data aggregation calculations on the sub-layer data in real time and dynamically in the background. The aggregation results and the affected element set are then pushed to the front end in an instant, enabling synchronized refreshing and interactive visualization of the map view and UI components (information panels, charts).

[0018] The entire process is transparent to the user, has a fast response time, and does not generate any static intermediate files.

[0019] A context-aware method for dynamic aggregation and linkage visualization of layers includes the following steps: Step 1: Construct a master-slave layer linkage model, including the following steps: Specify one main layer and at least one subordinate layer; Establish linkage rules between the main layer and the subordinate layers; Define a data aggregation operation, which is used to perform aggregation calculations on selected features from the layer points; Define how to visualize the results of data aggregation; Specifically: Main Layer Specification: Select a layer from the system's available layers as the trigger layer for interaction. For example, the project layer, park layer, subway layer, etc.

[0020] Specify from layers: Select one or more data layers that will be dynamically aggregated, such as a population layer or a corporate layer.

[0021] Spatial relationship definition, that is, to build the linkage rules between the main layer and the subordinate layer: define how the main layer elements establish spatial relationships with the subordinate layer elements.

[0022] This can be a fixed buffer, such as "500 meters around the main layer point feature"; or it can be a topological relationship, such as "intersecting with the main layer point feature" or "completely contained within the main layer point feature".

[0023] Data aggregation operation definition: Defines what calculations are performed on the selected set of features from the layer points.

[0024] For example, perform a count on the population layer to get the population number; perform a summation on the "Registered Capital" field of the corporate layer to get the total registered capital; or perform average (AVG), maximum (MAX), etc.

[0025] Visual template specification: Defines how the aggregation results are presented in the UI. Examples include "Display the text 'Service population: {results} people' in the panel with ID 'info-panel-population'", and "Highlight aggregated features from the layer in red on the map".

[0026] All configured linkage rules are stored in the system's linkage rule library.

[0027] In this embodiment, the main layer is the park layer, and the secondary layer is the population layer.

[0028] The linkage rule is: within 500 meters of the main layer point element.

[0029] In addition, when a layer (such as a population layer) is configured as a "from layer" in step one, the system performs a one-time preprocessing on it in the background. The steps for preprocessing point features in the from layer are as follows: Step 1A: Construct a memory-based spatial index; Step 1B: Construct Geohash encoding: Iterate through each point feature in the layer and obtain its latitude and longitude coordinates; Then, the Geohash algorithm is used to encode the two-dimensional latitude and longitude coordinates of each point feature into a one-dimensional string with a specified precision (e.g., a precision of 9). Step 1C: Construct an indexed in-memory data structure: The Geohash codes of all point features and their corresponding IDs are stored in an in-memory data structure that supports prefix matching.

[0030] Indexed in-memory data structures are hash tables (HashMap) or tries (Trie).

[0031] Step 2: Capture user interaction events and context.

[0032] Among them, user interaction events are mouse behavior events when a user clicks on an element on the main layer, including: mouse onClick (click) or onHover (hover) events.

[0033] The context is the information of the clicked feature on the main layer. The information of the clicked feature on the main layer includes: ID, geometric information, and attribute information.

[0034] In this embodiment, the main layer is the park layer, and the attribute information of the clicked feature on the main layer is: park name.

[0035] After the front end captures user interaction events and context, it immediately sends them to the system back end via API request or executes the real-time aggregation computing engine in the front end Web Worker, i.e., executes step 3.

[0036] Specifically: The interactive map (map module, i.e. the main layer) on the front end of the system continuously listens for user interaction events that are configured as the main layer, mainly mouse onClick or onHover events.

[0037] When a user clicks on a specific feature on the main layer (e.g., clicking on "Park X"), the interaction listener module (event listener) is triggered. The system immediately captures two key pieces of information: (1) Event type: onClick.

[0038] (2) Event context: The unique identifier (ID) of the clicked element and its complete geometric information (such as the polygon coordinate string of “a certain park”) and attribute information.

[0039] Among them, "event context" is... Figure 2 In this context, "spatial context" is referred to as "context".

[0040] This "context" is the core input for subsequent real-time computation.

[0041] The polygon coordinate string is the complete geometric representation of the clicked main layer feature, used in step 3B1, in "Query Range Determined": The system calculates an expanded query range based on the linkage rules (such as a "500-meter buffer distance") and the polygon coordinate string. The geometric boundary of this query range will be used in the subsequent step 3B2 to calculate the minimum bounding rectangle.

[0042] Attribute information: Key business attribute fields corresponding to this feature. For example, for the "Park Layer", attribute information may include: name, year, etc. Step 3a: Sometimes, linkage rules are not only based on geometric relationships, but also on attribute conditions.

[0043] The reason why point features can contain ID, geometric information, and attribute information: The main layer is the GIS system. The GIS system maintains three basic components for each spatial feature: a unique ID, a data value, and business attributes. This is the standard structure of a GIS data model.

[0044] Step 3, Dynamic Aggregation Calculation: Step 3A: Match the linkage rules based on the context.

[0045] Based on the context ID, the corresponding linkage rule is searched from the linkage rule library.

[0046] Step 3B: Filter the point features in the layer according to the linkage rules to obtain the set of features in the layer, which is the filtering result.

[0047] Among them, the information of the layer point features in the layer feature set includes: ID, attribute information and data value.

[0048] In this implementation, if the layer is the population layer, then: From the attribute information of the layer point features: population; Data value from layer point features: population.

[0049] Step 3B specifically involves the following steps: Step 3B1: Determine the query range.

[0050] Based on the event context (such as the polygon for "a certain park") and the linkage rules (such as "surrounding 500 meters"), the spatial range of this query is calculated in memory. Calculating the spatial range of the query is a prior art technique.

[0051] Step 3B2: Obtain the Geohash encoding prefix of the query range.

[0052] Calculate the minimum bounding box of the query range; Then, based on the required query precision, the minimum bounding rectangle is decomposed into a set of Geohash-encoded prefixes that can completely cover it; Step 3B3: Filter from the collection of layer features.

[0053] Using the Geohash encoding prefixes in the Geohash encoding prefix set, a query is performed in the memory index built in step 1C to retrieve all Geohash codes containing the Geohash encoding prefixes. The point features corresponding to all Geohash code IDs are thus formed from the layer feature set.

[0054] Specifically: The engine does not generate any static buffer layers. Instead, it performs a transient spatial lookup in memory based on the spatial relationships defined in the rules.

[0055] For example, the engine will traverse all point features from the layer (population layer) and determine in real time whether each point falls within a virtual range of 500 meters outward from the "Park" polygon.

[0056] All point features that meet the criteria are dynamically filtered out, forming a temporary set of features, i.e., the filtering result.

[0057] Step 3C: Perform data aggregation on the filtered results to obtain the aggregated results.

[0058] The engine performs the data aggregation operation defined in the rules on the temporary set of elements selected in the previous step.

[0059] For example, performing a COUNT operation on the set yields an aggregated result, such as the value 12345.

[0060] This computation is very lightweight, typically completed in milliseconds.

[0061] Step 4: Visualize the aggregation results according to the visualization method.

[0062] The calculation engine returns the aggregated results (values ​​12345) and the list of IDs of the filtered layer point features as response data to the front end.

[0063] Once the front-end receives the data, it immediately drives multiple views to update synchronously: Map view update: The linked visualization rendering module (map rendering engine) changes the display style of these population point elements in real time based on the returned ID list (for example, changing the color from gray to bright red, or increasing its radius), allowing users to intuitively see the affected range and objects in space.

[0064] These highlight styles will automatically revert when a user clicks on a blank area of ​​the map or an element in another main layer.

[0065] UI component view update: The data binding mechanism of the front-end framework (such as Vue or React) is triggered.

[0066] (1) Info Panel: The predefined panel content is dynamically updated to directly display the calculation results. For example, the panel text becomes: "Population served by this park: 12,345 people".

[0067] (2) Chart: If there are related charts (such as ECharts instances), their data sources will also be updated. For example, in a bar chart showing the population served by each park, the height of the bar corresponding to "Park X" will change dynamically to reflect the latest calculation results.

[0068] This concludes a complete context-aware, dynamic aggregation and linkage visualization process.

[0069] The selected list of IDs for layer point features is obtained from step 3B.

[0070] like Figure 1 As shown, a context-aware layer dynamic aggregation and linkage visualization system is used to implement a context-aware layer dynamic aggregation and linkage visualization method, including: The master-slave layer linkage model configuration module is responsible for implementing the functions in step 1. It provides a user interface for creating, reading, updating, and deleting linkage rules between master and slave layers, and persistently storing the rules in a database or configuration file.

[0071] The master-slave layer linkage model configuration module is stored in the "Data and Configuration Layer".

[0072] Figure 2 The data in "Read Data" refers to the information of point features on the layer.

[0073] Interactive listening module: corresponding to Figure 1 The client-side interaction layer in the system architecture is typically embedded in the front-end code of the map engine in a web browser or mobile app. It specifically listens for interaction events on the main map layer, captures the interaction context, and acts as the initiator of data requests.

[0074] Real-time aggregation computing engine: Responsible for implementing the functionality in step 3. It is the core backend of the system, receiving frontend requests, performing efficient in-memory filtering and data aggregation calculations, and quickly returning results. This engine is designed to be stateless to support high concurrency.

[0075] Linked Visual Rendering Module: Responsible for implementing the functionality in step 4. It is the front-end data-driven view layer, containing map rendering logic and UI components. It subscribes to the results returned by the calculation engine and is responsible for updating the map feature styles and the display content of various UI components in a linked manner.

[0076] For some implementation steps of the technical solution of this invention, there are feasible alternatives. These alternatives can optimize the invention or provide diversified functions in different scenarios: 1. Alternatives to interactive triggering methods: Hover Trigger: In addition to click triggers (onClick), mouse hover (onHover) can also be used as an event trigger. When the user hovers the mouse over an element in the main layer, the system dynamically aggregates and links the visualizations; the view returns to its original state when the mouse moves away. This method provides a more lightweight "preview" function and a smoother interaction.

[0077] Custom-range triggering: Users are not limited to clicking on a single feature; they can also use the "pen" or "lasso" tools to freely draw a polygon on the map. This custom polygon becomes a dynamic "context," and the system will aggregate the sub-layer data within its range in real time.

[0078] 2. Alternatives to data aggregation operations: This solution lists basic aggregation operations such as COUNT and SUM. Alternative solutions can support more complex custom aggregation scripts. For example, administrators can write short JavaScript or Python scripts to perform weighted averages, spatial density calculations, or more complex business metric statistics, greatly enhancing the flexibility of aggregation calculations.

[0079] 3. Alternatives to computing engine deployment methods: In this solution, the computing engine can be deployed on a backend server. An alternative is to utilize WebAssembly technology to deploy some or all of its logic in a frontend Web Worker. This reduces server load, lowers network latency, and provides faster response times for medium-sized datasets, making it particularly suitable for purely frontend or offline applications.

[0080] 4. Substitution of spatial relationships: Besides buffers and topological relationships, spatial relationships can also be based on network distance. For example, if the main layer is "fire station" and the secondary layer is "residential building," the relationship could be "reachable within 10 minutes along the road network," which is closer to practical applications than a simple linear buffer. This requires the system to integrate road network data and network analysis capabilities.

[0081] Compared with the prior art, the present invention has the following significant advantages: 1. Extremely high interactive efficiency and immediacy: The complex analysis process in traditional GIS, which requires multiple steps and takes minutes, is simplified to a single click or hover action by the user, with the system responding in milliseconds. Users do not need professional knowledge to conduct in-depth data exploration, achieving true "click and get results".

[0082] 2. Intelligent Context Awareness: Through a master-slave layer linkage model, the system can "understand" the user's interaction intent. Every user interaction becomes a dynamic "analysis context," driving the system to perform targeted, real-time calculations, rather than passively executing general tools.

[0083] 3. Intuitive and immersive multi-view interactive visualization: The aggregated results are not presented as isolated layers or tables, but are reflected synchronously and interactively in multiple views such as map space highlighting, information panel value updates, and dynamic chart changes. This integrated presentation method greatly enhances the interpretability and insight of the data.

[0084] 4. Significantly Reduced System Resource Consumption: Core computation processes are completed dynamically in memory, avoiding the generation of any static or temporary intermediate layers or data files. This greatly saves disk I / O and storage space, and reduces memory usage. This allows the system to maintain high performance and stability even when processing massive amounts of data and high-concurrency requests.

[0085] 5. Excellent scalability and flexibility: The configuration of linkage rules is modular, and business personnel can add new layer linkage relationships at any time according to their needs (for example, using "subway station" as the main layer and associating it with "office building" as the secondary layer to calculate its average rental price) without modifying the core system code, which has strong business adaptability and scalability.

[0086] 6. Superior performance and low resource consumption under massive data: The innovative in-memory spatial indexing algorithm based on Geohash fundamentally avoids brute-force traversal of the entire dataset. It reduces the computational complexity of large-scale spatial queries from O(N) to close to O(logN) or O(1), ensuring that the system can maintain millisecond-level response when processing millions or even tens of millions of data points. At the same time, by avoiding the generation of static intermediate files, it greatly saves disk I / O and storage space, and significantly reduces the overall resource consumption of the system.

[0087] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the present invention. Any equivalent substitutions or modifications made by those skilled in the art to the technical solutions and content disclosed in the present invention without departing from the scope of the present invention shall be deemed to have remained within the protection scope of the present invention.

Claims

1. A method for context-aware based dynamic aggregation and linkage visualization of layers, characterized in that, The method comprises the following steps: Step 1, constructing a master-slave layer linkage model, comprising the following steps: designating a master layer and at least one slave layer; constructing linkage rules between the master layer and the slave layer; defining a data aggregation operation for aggregating and calculating selected point elements of the slave layer; defining a visualization mode of the data aggregation result; Step 2, capturing a user interaction event and a context; wherein the user interaction event is a mouse behavior of the user when clicking on an element on the master layer; and the context is information of the clicked point element on the master layer; Step 3, dynamic aggregation calculation: Step 3A, matching the linkage rules according to the context; Step 3B, filtering point elements in the slave layer according to the linkage rules to obtain a slave layer element set, i.e. a filtering result; Step 3C, performing a data aggregation operation on the filtering result to obtain an aggregation result; Step 4, visualizing the aggregation result according to the visualization mode.

2. The context-aware based layer dynamic aggregation and linkage visualization method according to claim 1, wherein, The linkage rules comprise one of a set distance around the clicked element of the master layer and a topological relationship between the clicked element of the master layer.

3. The context-aware based layer dynamic aggregation and linkage visualization method of claim 1, wherein, The information of the clicked point element on the master layer comprises an ID, geometric information and attribute information.

4. The context-aware based layer dynamic aggregation and linkage visualization method of claim 1, wherein, The information of the slave layer point element in the slave layer element set comprises an ID, attribute information and a data value.

5. The context-aware based layer dynamic aggregation and linkage visualization method of claim 1, wherein, Step 1 further comprises a step of data preprocessing of point elements in the slave layer: Step 1A, constructing an in-memory spatial index; Step 1B, constructing a Geohash code: traversing each point element in the slave layer to obtain its latitude and longitude coordinates; then, using a Geohash algorithm to encode the two-dimensional latitude and longitude coordinates of each point element into a one-dimensional string with a specified precision; Step 1C, constructing an indexed in-memory data structure: storing the Geohash codes of all point elements and their corresponding IDs in an in-memory data structure supporting prefix matching.

6. The context-aware based layer dynamic aggregation and linkage visualization method of claim 5, wherein, Step 3B specifically comprises the following steps: Step 3B1, query range determination: calculating the spatial range of this query in memory according to the event context and the linkage rules; Step 3B2, obtaining the Geohash code prefix of the query range: calculating the minimum bounding box of the query range; then, according to the required query precision, decomposing the minimum bounding box into a set of Geohash code prefixes that can completely cover it; Step 3B3, filtering the slave layer element set: using the Geohash code prefixes in the Geohash code prefix set to query the in-memory index constructed in Step 1C, retrieving all Geohash codes containing the Geohash code prefixes, and all point elements corresponding to the Geohash code IDs, i.e. forming the slave layer element set.

7. The context-aware based layer dynamic aggregation and linkage visualization method of claim 5, wherein, The indexed in-memory data structure is a hash table HashMap or a prefix tree Trie.

8. A context-aware based layer dynamic aggregation and linkage visualization system for implementing the context-aware based layer dynamic aggregation and linkage visualization method of any one of claims 1-7, characterized in that, It comprises: a master-slave layer linkage model configuration module; and an interaction listening module, a real-time aggregation calculation engine and a linkage visualization rendering module connected in sequence; one input end of the interaction listening module is connected to the output end of the master layer; one input end of the real-time aggregation calculation engine is connected to the output end of the master-slave layer linkage model configuration module.

Images